Clinical Microbiology and Infection 28 (2022) 1288.e1e1288.

e7

Contents lists available at ScienceDirect

Clinical Microbiology and Infection

journal homepage: www.clinicalmicrobiologyandinfection.com

Original article

Multiple colony antifungal susceptibility testing detects polyresistance

in clinical Candida cultures: a European Confederation of Medical
Mycology excellence centers study
Miriam A. Knoll 1, Nina Lackner 1, Hanno Ulmer 2, Eldina Samardzic 1, Joerg Steinmann 3, 4,
Robert Krause 5, 6, Hedda L. Verhasselt 4, Peter-Michael Rath 4, Frieder Fuchs 7, 8,
Philipp Koehler 9, 10, Blandine Denis 11, Samia Hamane 11, 12, Alexandre Alanio 12, 13,
€ rl 1, *
Cornelia Lass-Flo
1)
Institute of Hygiene and Medical Microbiology, ECMM Excellence Center, Medical University of Innsbruck, Innsbruck, Austria
2)
Department of Medical Statistics, Informatics and Health Economics, Medical University of Innsbruck, Innsbruck, Austria
3)
Institute of Clinical Hygiene, Medical Microbiology and Infectiology, Paracelsus Medical University, Klinikum, Nurnberg, Germany
4)
Institute of Medical Microbiology, ECMM Excellence Center, University Hospital Essen, University of Duisburg-Essen, Essen, Germany
5)
Biotechmed, Graz, Austria
6)
Division of Infectious Diseases, ECMM Excellence Center, Department of Internal Medicine, Medical University of Graz, Austria
7)
Institute for Medical Microbiology, Immunology and Hygiene, University of Cologne, Medical Faculty and University Hospital of Cologne, Cologne,
Germany
8)
Department of Microbiology and Hospital Hygiene, Bundeswehr Central Hospital Koblenz, Koblenz, Germany
9)
Department I of Internal Medicine, ECMM Excellence Center, Critical Care Medicine, University Hospital of Cologne, Cologne, Germany
10)
Cologne Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Cologne, Germany
11)
Infectious Diseases Department, ECMM Excellence Center, Ho ^pital Saint-Louis, APHP, Paris, France
12)
Mycology-Parasitology Department, Ho ^pital Saint-Louis, APHP, Paris, France
13)
Universite de Paris, Molecular Mycology Unit, CNRS UMR2000, National Reference Centre for Invasive Mycoses and Antifungals, Institut Pasteur,
Universit e de Paris, Paris, France

a r t i c l e i n f o a b s t r a c t

Article history:
Objectives: Many factors influence the outcome of in vitro antifungal susceptibility testing (AFST),
Received 7 February 2022
including endpoint definition, inoculum sizes, time and temperature of incubation, and growth medium
Received in revised form
13 April 2022 used. This European Confederation of Medical Mycology (ECMM) Excellence center driven study
Accepted 16 April 2022 investigated multiple colony testing (MCT) of five separate colonies to investigate the prevalence of
Available online 9 May 2022 polyresistance (PR), defined as heterogeneous MICs from a same-species Candida culture irrespective of
the underlying resistance mechanism.
Editor: E. Roilides Methods: Candida spp. MCT for fluconazole and anidulafungin was performed by Etest prospectively
comprising 405 clinical samples. MCT results were compared to the real-life routine MIC data and PR was
Keywords: assessed. Candida colonies displaying strong PR were selected for genotyping using multilocus sequence
Antifungal resistance typing and random amplified polymorphic DNA assays for C. lusitaniae.
Antifungal susceptibility testing
Results: Candida PR was observed in 33 of 405 samples (8.1%), with higher rates for non-albicans species
Candida
(26/186, 14%) than for C. albicans (7/219, 3.2%), and for fluconazole than for anidulafungin. MCT detected
Multiple colony testing
Polyresistance acquired resistance more often than routine AFST (18/405, 4.5%) and 9 of the 161 investigated blood
cultures showed PR (5.6%). Multilocus sequence typing and random amplified polymorphic DNA did not
reveal a uniform genetic correlate in strains studied.
Conclusions: This study shows that Candida single MIC-values obtained in routine diagnostics may be
incidental, as they fail to detect PR and resistant subpopulations reliably. The reasons for PR seem to be
manifold and should be regarded as a phenotypical expression of genomic variability irrespective of the

* Corresponding author. Cornelia Lass-Flo€rl, Institute of Hygiene and Medical Microbiology, Medical University of Innsbruck, 6020 Innsbruck, Austria.
€ rl).
E-mail address: cornelia.lass-floerl@i-med.ac.at (C. Lass-Flo

https://doi.org/10.1016/j.cmi.2022.04.014
1198-743X/© 2022 Published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious Diseases. This is an open access article under the CC BY
license (http://creativecommons.org/licenses/by/4.0/).
 M.A. Knoll et al. / Clinical Microbiology and Infection 28 (2022) 1288.e1e1288.e7
underlying resistance mechanism, which may help to interpret ambiguous and non-reproducible AFST
results. Miriam A. Knoll, Clin Microbiol Infect 2022;28:1288.e1e1288.e7
© 2022 Published by Elsevier Ltd on behalf of European Society of Clinical Microbiology and Infectious
Diseases. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/
Introduction
Invasive Candida infections are an emerging threat with high
mortality and morbidity, particularly in immunocompromised pa-
tients [1]. Due to an increase of drug resistant species, monitoring
and detecting antifungal resistance is important [2]. Multiple col-
ony testing (MCT) for antifungal susceptibility testing (AFST) of
Aspergillus spp. is recommended [3] as azole-resistant and -sus-
ceptible Aspergillus strains may be simultaneously present in-
patient samples [4]. MCT aims to detect resistant subpopulations,
which otherwise may lead to breakthrough fungal infections [5,6].
Similarly, MCT for Candida samples has been suggested studying its
within-host genomic diversity [7]. Next generationemultilocus
sequence typing showed that 12% of patient samples contained
two unrelated Candida strains [8]. In line, longitudinal samples
from oral candidiasis showed several genetic and phenotypical
changes throughout antifungal treatment, including different co-
existing MICs for fluconazole [9].
To ensure the inclusion of intra-sample heterogeneity, European
Committee on Antimicrobial Susceptibility Testing (EUCAST)
reference method recommends to pick up five representative col-
onies for inoculum suspension [10]; by contrast, the more
commonly used Etest method (gradient strip test) recommends
using ‘a few well-isolated colonies” [11]. Overall, it is rather coin-
cidental how many colonies are truly used in routine AFST, and
improbable that a presumed intra-sample heterogeneity can be
displayed by one test per sample.
The aim of this study was to assess the prevalence of intra-
sample variance of AFST in a variety of usually sterile specimens
and Candida species, applying MCT in a prospective multi-center
study across Europe. We defined here the term polyresistance
(PR) as heterogeneous MICs from a same-species Candida culture.
Material and methods
Multiple colony testing (MCT)
The frequency of PR within taxonomically identical strains was
investigated by separately testing five distinct colonies from each
clinical culture. Centers from Austria (2), France (1), and Germany
(3) participated, six of them being European Confederation of
Medical Mycology (ECMM) Excellence centers (ECMM EC). From
March 2020 to March 2021, we prospectively collected 405 clinical
samples from sterile regions with growth of Candida spp. Mixed
yeast infections were separately tested, and no consecutive isolates
were used. From each of the original samples, five morphologically
identical colonies labeled AeE were separately subcultured on a
chromogenic agar for Candida (bioMe rieux, Marcy l’Etoile, France
and BBL, Becton Dickinson GmbH, Heidelberg, Germany) to ensure
purity. Species identification was performed for each subculture
(AeE) by Matrix-Assisted Laser Desorption/Ionization Time of
Flight Mass Spectrometry (MALDI-TOF MS) Biotyper (Bruker Dal-
tonik GmbH, Bremen, Germany). AFST with fluconazole and ani-
dulafungin was performed separately for each of the five pure
subcultures (AeE), using the Etest gradient diffusion method ac-
cording to the manufacturer's instructions [11]. Etest results were
1288.e2
4.0/).
translated to EUCAST breakpoints to categorize the MICs [12].
Categorization for species without EUCAST breakpoints was
applied as described elsewhere [13e16].
The results of each test (AeE) were compared to the corre-
sponding other colonies and to the MICs obtained in routine
diagnostic assays (Etest (bioMe rieux), MICRONAUT-AM (MERLIN
Diagnostika GmbH, Bornheim, Germany) and Vitek 2 AST-YS 08
card (bioMerieux)). Differences in categorization and above three
log2 dilutions were documented as PR. PR was classified in the
following three categories: categorically (EUCAST S, I, R) poly-
resistant MICs above three log2 dilutions (strong PR), categorically
(EUCAST S, I, R) polyresistant MICs up to three log2 dilutions
(moderate PR), and polyresistant MICs above three log2 dilutions
without categorical (EUCAST S, I, R) difference (weak PR). Isolates
with PR were separately cryopreserved and collected at Medical
University of Innsbruck for further analysis, if indicated.
Reproducibility and genotyping
For selected polyresistant samples, repetition of AFST by
EUCAST microdilution and Etest was performed [11,12]. Four sam-
ples displaying strong PR were selected for genotyping of the five
separate isolates. DNA was extracted from fresh culture using Type
C bead tubes (Macherey-Nagel, Germany) and the Monarch
genomic DNA purification kit T3010 (New England Biolabs, Ipswich,
MA, USA), according to the manufacturer's instructions [17].
Multilocus sequence typing (MLST) was performed for
C. albicans, C. glabrata, and C. tropicalis strains according to the
recommendations on https://pubmlst.org [18] with primers pub-
lished elsewhere (Table S1) [19e21]. The PCR products were sent to
Eurofins Genomics (Ebersberg, Germany) for Sanger sequencing.
Sequencing chromatograms were analyzed in Sequence Scanner
(Version 2.0, ThermoFisher Scientific, Waltham, MA, USA) and
manually corrected for heterologous sites. MLST profiles were
determined using the online tools on https://pubmlst.org [18].
In lack of a MLST scheme for C. lusitaniae, we performed random
amplified polymorphic DNA (RAPD) for these strains using three
random primers OPA-1, OPA-8, and OPA-10 (Table S2) [22,23]. PCR
products were separated in a 1.5% agarose gel at 100 V for 3 h and
analyzed with the Gel DocTM EZ Imager and the Image Lab soft-
ware Version 5.2.1 (both Bio-Rad Laboratories, Hercules, CA, USA).
Statistical analysis
The sample size was prespecified with 400 samples to have 80%
power to detect acquired resistances of at least 3% between MCT
and routine AFST as statistically significant using McNemar test.
McNemar test was used to compare detection rates within the
same samples between different methods. Pearson chi-square test
was applied to compare detection rates between samples of
different species. P < 0.05 were considered to indicate statistical
significance. Statistical analyses were performed with IBM SPSS
Statistics 26 (IBM Corp., Armonk, NY, USA).
This project was part of a laboratory quality management study
within ECMM Excellence Centres; the study was approved by the
hospital's ethics committee (Study no. 1166/2018).
1288.e3 M.A. Knoll et al. / Clinical Microbiology and Infection 28 (2022) 1288.e1e1288.e7

Results
MCT for fluconazole and anidulafungin was performed for 405
clinical samples with Candida spp. in six centers. Sample origins are
summarized in Table 1, MICs from routine diagnostic assays are
summarized in Table S3.
Altogether, PR of any category was obtained in 8.1% (33/405) of
all Candida culture samples (Table 2). Predominantly PR was
moderate, meaning categorically (EUCAST S, I, R) polyresistant MICs
within three log2 dilutions, with often borderline R MIC values. PR
was significantly more frequent in non-albicans species than in
C. albicans (26/186, 14.0% vs. 7/219, 3.2%; p < 0.001). There was a
tendency towards higher PR rates for fluconazole than for anidu-
lafungin; however, this trend was not statistically significant (22/
405, 5.4% vs. 13/405, 3.2%; p ¼ 0.08). In C. albicans isolates with
fluconazole MICs above 1 mg/L, 3/7 showed moderate PR in MCT. In
C. glabrata, 8/11 PR samples showed fluconazole MICs above 8 mg/L
in routine AFST, while reversely, 8/43 samples with fluconazole
MICs above 8 mg/L showed PR upon re-testing. After exclusion of
species with clinical breakpoints above 0.064 mg/L for anidula-
fungin, 9/21 samples with high MICs above 0.016 mg/L showed PR in
MCT (43%). To estimate the implications of PR, differences between
routine results and MCT were labeled as very major error (VME, S
vs. R), major error (ME, R vs. S), minor error (mie, S vs. I or I vs. R),
and very minor error (vmie, I vs. S or R vs. I). MCT revealed 6 VMEs,
5 ME, 12 mie, and 5 vmie (Table 3). Acquired resistances for flu-
conazole were significantly more often detected by MCT than by
routine AFST (34/405, 8.4% vs. 25/405, 6.2%; p ¼ 0.012), while for
anidulafungin, this was not statistically significant (12/405, 3.0% vs.
9/405, 2.2%, MCT vs. routine AFST, respectively; p ¼ 0.453).
Four samples with strong PR were selected for genotyping of the
five separate isolates, MICs with replicates are displayed in Table 4.
MLST was performed for samples I 38 (C. albicans), I 15 (C. glabrata),
and E 23 (C. tropicalis). In the absence of an MLST scheme for
C. lusitaniae, RAPD was performed for sample K 58.
In sample I 38 (C. albicans), isolate E displayed an elevated MIC
for fluconazole (4 mg/L) compared to the other isolates (0.25e0.5 mg/
L). The obtained MLST sequences allowed unambiguous determi-
nation of the sequence types (ST 1110), which was identical between
the isolates AeD (Table 4). For isolate E, however, no clean sequence
could be obtained for the locus SYA1. All other loci were identical
with the other isolates. In sample I 15 (C. glabrata), isolates B and D
displayed an I MIC for fluconazole of 4 and 8 mg/L, the same as ob-
tained in routine AFST. In contrast, isolates A, C and E displayed MICs
>256 in Etest. Upon repetition by Etest and EUCAST microdilution,
isolate D showed insufficient growth for microdilution, while the
remaining tests showed partly fluctuating MICs. MLST schemes for
all isolates were identical (ST 3), except for B, for which no clean
sequence could be obtained for the locus TRP1 (Table 4). Sample E
23 (C. tropicalis) showed MICs for fluconazole in the S, I and R
category. However, upon repetition with Etest and EUCAST micro-
dilution, all MICs were in the S category (0.125-0.25 mg/L). All iso-
lates showed identical MLST profiles, however, without an existing
match in the database (Table 4). In sample K 58, isolate A had shown
macrocolonies for fluconazole Etest, which upon repetition showed
a MIC of 4 mg/L in Etest and 8 mg/L in EUCAST microdilution. With
primers OPA-1 and OPA-8, RAPD bands differed between isolate A
and BeD (Fig. 1).
Altogether, PR occurred in up to 8.1% (33/405) of samples
replicated by MCT, with higher rates for non-albicans species than
for C. albicans. Strong PR was observed in 2% (8/405) of samples,
again with higher rates in non-albicans species. Genotyping of
selected isolates showed that the reasons for PR were manifold.
Discussion
PR, here defined as heterogeneous results in AFST of five different
colonies from a same-species Candida sample, was observed in 33 of
405 samples (8.1%). Under ideal conditions to reduce biological and
methodological variation, repeated testing of the same Candida wild-
typ (WT) isolate produced MIC values within a three log2 dilution
range [24]. Hence, we defined a three log2 dilution range within the
same EUCAST category as an accepted technical variance. MICs
exceeding this range or differences in categories (EUCAST S, I, R or
WT, non-WT) were defined as PR. Strong PR with categorically
(EUCAST S, I, R) different MICs above three log2 dilutions is clearly
the most pronounced form of PR and occurred with a rate of 2%
across all species. In clinical practice, moderate PR may have the
same implications for patient management as strong PR, despite
smaller differences in MICs. Contrarily, weak PR may not influence
patient management, but nevertheless questions the reliability of
AFST. Moderate PR with borderline MICs in different EUCAST cate-
gories less than three log2 dilutions apart, was more common for
fluconazole than for anidulafungin, especially with C. glabrata vary-
ing between I and R. Generally, the good agreement between the two
methods allows translation of Etest MICs to EUCAST breakpoints
[25,26]. However, due to some methodological variance, epidemio-
logical cut-offs (ECVs) specifically evaluated for Etest may more
sensitively identify resistances than reference break-points (BPs)
evaluated for microdilution [27,28]. For both fluconazole and ani-
dulafungin, ECVs for Etest are usually lower than EUCAST clinical
breakpoints [28,29]. Especially for low-level resistances, MCT may

Table 1
Sample origins and number of samples tested with multiple colony testing per center

Sample type On-site centers Central laboratory Total

Essen Graz Cologne Nürnberg Paris Innsbruck

Blood cultures 33 8 66 30 10 14 161

Urine 34 44 78
Intraabdominal 9 4 31 33 77
Bile 4 4 18 26
Pleural/lung 6 1 7
Catheter tip 2 1 8 11
Tissue 9 7 3 19
Others 15 2 3 6 26
Total 100 20 66 82 10 127 405
PR samples 15 1 4 5 0 8 33
Samples selected for genotyping 1 0 1 0 0 2 4

Other types of samples included puncture and drainage fluids from various sterile regions, swabs from sterile regions, gastric liquid, and wound secretions. Polyresistant
samples from on-site centers were sent to the central laboratory, where genotyping was performed for selected samples.
AbbreviationsPR, prevalence of polyresistance.
 M.A. Knoll et al. / Clinical Microbiology and Infection 28 (2022) 1288.e1e1288.e7 1288.e4

Table 2
Number of PR samples as obtained by multiple colony testing and PR samples by species in percent

Species Polyresistant results Homo-genous results

Strong PR Moderate PR Weak PR Total PR

FLU ANI Total FLU ANI Total FLU ANI Total

Candida albicans (n ¼ 219) 2a 2 3 3a 6 7a 212

C. glabrata (n ¼ 91) 1 1 9 1 10 1 2 3 14 77
C. parapsilosis (n ¼ 25) 1 1 2 2 4 5 20
C. tropicalis (n ¼ 23) 2 1 3 1a 1a 1a 4a 19
C. krusei (n ¼ 13) 2 2 2 11
C. lusitaniae (n ¼ 2) 1 1 1 1
Other species 0 32
Total, n 6 2 8 15 7 21a 1 4 5 33a 372
C. albicans, % (n ¼ 219) 0.9 0.0 0.9 1.4 1.4 2.7 0.0 0.0 0.0 3.2 96.8
non-albicans, % (n ¼ 186) 2.2 1.1 3.2 6.5 2.2 8.6 0.5 2.2 2.7 14.0 86.0
Total, % 1.5 0.5 2.0 3.5 1.7 5.2 0.2 1.0 1.2 8.1 91.9

PR was defined as heterogeneous antifungal susceptibility testing results from a same-species Candida sample upon re-testing of colonies. PR was divided in categorically
(EUCAST S, I, R) polyresistant MICs above 3 log2 dilutions (strong PR), categorically (EUCAST S, I, R) polyresistant MICs up to 3 log2 dilutions (moderate PR) and polyresistant
MICs above 3 log2 dilutions without categorical (EUCAST S, I, R) difference (weak PR). Species with only homogeneous results by multiple colony testing were summarized as
other species. Other species included C. dubliniensis (n ¼ 16), C. kefyr (n ¼ 7), C. fabianii (n ¼ 2), C. inconspicua (n ¼ 2), Saccharomyces cerevisiae (n ¼ 2), C. metapsilosis (n ¼ 1),
C. auris (n ¼ 1), and C. guilliermondii (n ¼ 1).
a
Same isolate exhibiting polyresistant results for FLU and ANI.
Abbreviations: ANI, anidulafungin; EUCAST, European Committee on Antimicrobial Susceptibility Testing; FLU, fluconazole; PR, polyresistance

Table 3 occurred in our study with higher rates in non-albicans species than
Polyresistant isolates obtained by multiple colony testing listed by the origin of the in C. albicans, and more frequently for fluconazole than for anidu-
samples
lafungin. These differences must likely be considered as a product
Specimen type PR Number of polyresistant isolates of overall resistance rates, which determine the probability of
Candida albicans Non-albicans encountering resistant subpopulations. These insights may aid in
the consideration of ambiguous AFST results in non-albicans spe-
FLU ANI FLU ANI
cies and treatment with fluconazole.
Blood cultures (n ¼ 161) Strong 0 0 1 1 The presence of multiple Candida strains was more commonly
Moderate 0 0 5 2
found in feces, urine, and vaginal samples than in sterile body sites
Weak 0 0 0 0
Intraabdominal (n ¼ 77) Strong 0 0 1 0 [7e9], supporting the hypothesis that only a small number of cells
Moderate 0 1 2 1 enter a sterile site of infection [8]. However, our PR samples
Weak 0 0 0 3 originated from a variety of infected sterile body sites, with urine
a
Urine (n ¼ 78) Strong 1 0 1 1
a a a
and bile samples as an exception, yet usually free from a colo-
Moderate 2 1 5 1
Weak 0 0 0 1
nizing Candida flora. Nine (5.6%) of the 161 investigated blood
Others (n ¼ 89) Strong 1 0 1 0 cultures were PR, demonstrating that PR as a manifestation of
Moderate 1 1 0 0 intra-host diversity also occurs in sterile body regions. Several
Weak 0 0 1 0 lineages may co-exist or rapidly adapt under stress [9], resulting in
PR was marked in categorically (EUCAST S, I, R) polyresistant MICs above three log2 PR upon separate testing of single colonies. Accordingly, cell-to-
dilutions (strong PR), polyresistant MICs up to three log2 dilutions (moderate PR), cell variations in drug response (heteroresistance) have been
and polyresistant MICs above three log2 dilutions without categorical (EUCAST S, I,
described for fungi, with high rates of heteroresistance for
R) difference (weak PR). Others included sterile puncture fluids, bile, tissues, gastric
liquids, wound secretions, and drainage fluids from various sterile regions. C. glabrata and fluconazole, however, not for echinocandins [30].
a
Same isolate exhibiting polyresistant results for FLU and ANI. Population analysis profiling [30] or the molecular analysis of
Abbreviations: ANI, anidulafungin; EUCAST, European Committee on Antimicrobial resistance mechanisms enable an elaborate characterization of
Susceptibility Testing; FLU, fluconazole; PR, polyresistance. the underlying reasons for some types of PR, however, these
methods are not suitable for routine application. The term PR here
was defined as heterogeneous MICs from a same-species Candida
aid disclosing otherwise undetected resistances, as some MICs will culture, to help describe heterogeneous AFST results irrespective
be read above the breakpoint concentration. Altogether, colonies of the underlying mechanism.
with a categorically (EUCAST S, I, R) higher MIC than in routine AFST In our study, we performed genotyping to assess the clonality of
were found in 4.4%. Classification of MICs close to the breakpoint four strains with the most pronounced strong PR to address the
concentration was less consistent upon re-testing, meaning that possibility of multiple infections of the same species but differently
some colonies showed MICs above and some below the breakpoint, susceptible strains. In MLST, two combinations of sample and locus
often resulting in moderate PR. However, for samples with single repeatedly failed to give clear sequencing results for unknown
deviant colonies, the MIC of the routine AFST result was not indic- reasons. Other studies demonstrated the co-existence of multiple
ative of PR. Routine AFST builds the basis for decisions regarding related but divergent genotypes differing in several thousand single
antifungal therapy, hence fast and reliable results are necessary. Our nucleotide polymorphisms (SNPs) and harboring different flucon-
study shows that in PR, AFST results are partly incidental, potentially azole MICs [9]. Recent data indicate that MLST only rarely reveals
leading to suboptimal selection of antifungal therapy or break- this diversity, namely when an SNP affects by chance one of the
through infections. MLST loci [7]. A similar background might be possible for our
Generally, non-albicans species show higher resistance rates C. albicans, C. glabrata, and C. tropicalis samples with largely iden-
than C. albicans, and acquired resistance occurs more frequently tical MLST loci. Underlying might be an unknown number of SNPs
towards azoles than towards echinocandins [2]. Accordingly, PR becoming apparent as PR but not discovered by MLST.
1288.e5 M.A. Knoll et al. / Clinical Microbiology and Infection 28 (2022) 1288.e1e1288.e7

Table 4
Multilocus sequence typing allele types and sequence types for individual isolates from samples with strong polyresistance

Candida albicans Alleles Sequencing MIC FLU MIC repetition Etest MIC EUCAST Reproducible
type microdilution
Isolate ID AAT1a ACC1 ADP1 MPIb SYA1 VPS13 ZWF1b

ATCC 90028 5 3 5 9 2 25 12 / n.a. n.a. n.a. n.a.

I 38 A 13 10 15 6 7 32 15 1110 0.25 0.25 0.125 yes
I 38 B 13 10 15 6 7 32 15 1110 0.5 0.125 0.125 yes
I 38 C 13 10 15 6 7 32 15 1110 0.5 0.25 0.125 yes
I 38 D 13 10 15 6 7 32 15 1110 0.5 0.125 0.125 yes
I 38 E 13 10 15 6 d 32 15 d 4 4 2 yes

C. glabrata Alleles Sequencing MIC FLU MIC Repetition Etest MIC EUCAST Reproducible
Isolate ID FKS LEU2 NMT1 TRP1 UGP1 URA3 type microdilution

ATCC 90030 8 4 3 5 1 2 10 n.a. n.a. n.a. n.a.

I 15 A 5 7 8 7 3 6 3 >256 64 32 no
I 15 B 5 7 8 / 3 6 d 4 8 8 yes
I 15 C 5 7 8 7 3 6 3 >256 128 32 no
I 15 D 5 7 8 7 3 6 3 8 8 x no
I 15 E 5 7 8 7 3 6 3 >256 128 32 no

C. tropicalis Alleles Sequencing MIC FLU MIC Repetition Etest MIC EUCAST Reproducible
Isolate ID ICL1 MDR1 SAPT2 SAPT4 XYR1 ZWF1a type microdilution

ATCC 90874 1 1 11 9 77 1 d n.a. n.a. n.a. n.a.

E 23 A 1 1 10 1 73 1 d 4 0.125 0.125 no
E 23 B 1 1 10 1 73 1 d 8 0.25 0.125 no
E 23 C 1 1 10 1 73 1 d 0.5 0.25 0.125 yes
E 23 D 1 1 10 1 73 1 d 4 0.125 0.125 no
E 23 E 1 1 10 1 73 1 d 8 0.125 0.25 no

MICs in mg/L obtained by multiple colony testing with Etest are displayed on the right and marked as reproducible within ±2 log2 dilutions by Etest and the EUCAST
microdilution or as not reproducible.
Abbreviations: EUCAST, European Committee on Antimicrobial Susceptibility Testing; FLU, fluconazole; n.a., not applicable.

The selected PR C. lusitaniae sample showed a reproducible MIC
of 4 mg/L for isolate A as opposed to 0.064 mg/L for the remaining
four isolates and the routine result. RAPD band patterns for isolate
A differed from isolates BeE, suggesting the presence of a second
C. lusitaniae clone in the sample. Altogether, the results of geno-
typing PR isolates indicate that the reasons for PR can be manifold,
with no single underlying mechanism.
There are some limitations to our study. Some of the R MCT
results were not confirmed by a second method. Especially for
borderline R results, the technical variance between Etest and
microdilution can produce different results with another method.
However, the aim of the study was to evaluate the level of variance
obtained by multiple testing in a real-life setting. Furthermore, we
did not investigate the underlying molecular resistance mecha-
nisms of the PR strains, as this would have exceeded the objective
of the study. This should be elucidated in a subsequent study.
Moreover, MLST as used in our study has lower discriminatory
power in genotyping Candida, than for example whole-genome
sequencing, which would detect all SNPs of diversifying lineages
[7]. Differences between diverging progeny of the same strain may
therefore stayed undiscovered.
In conclusion, our results demonstrate that PR is a real-life phe-
nomenon and that one single MIC-value obtained by standard AFST
may not be as reliable as expected. As a result, for clinical practice,
more than one colony should be included in AFST. Further research is
needed about the clinical implications of this phenomenon. How-
ever, we must ultimately assume that resistances may go undetected
by standard AFST, and consider applying adaptions such as MCT or
modified inocula to the repertoire of AFST.
Transparency declaration

Fig. 1. Band patterns obtained with primers OPA-1 and OPA-8 for C. lusitaniae. Bands
differed between isolate A (bands 1 and 6) and isolates BeD (bands 2e5 and 7e10) for
both primers. M, 1 kb marker.
The authors received no financial support for this work and have
declared no conflicts of interest exists. J.S.T. reports payment or
honoraria for lectures, presentations, speaker's bureaus, manu-
script writing or educational events, and leadership or fiduciary
role in other board, society, committee or advocacy group of Gilead.
R.K. received research grants from Merck and Pfizer and served on
the speakers' bureau of Pfizer, Gilead, Astellas, Basilea, Merck, and
Angelini. B.D. reports consulting fees from Pfizer and payment for
lectures, presentations, and support for attending meetings and/or
travel from Gilead. P.K. reports grants or contracts from German
Federal Ministry of Research and Education (BMBF) B-FAST (Bun-
desweites Forschungsnetz Angewandte Surveillance und Testung)
and NAPKON (Nationales Pandemie Kohorten Netz, German Na-
tional Pandemic Cohort Network) of the Network University
Medicine (NUM) and the State of North Rhine-Westphalia;
consulting fees Ambu GmbH, Gilead Sciences, Mundipharma
Research Limited, Noxxon N.V. and Pfizer Pharma; honoraria for
lectures from Akademie für Infektionsmedizin e.V., Ambu GmbH,
Astellas Pharma, BioRad Laboratories Inc., European Confederation
 M.A. Knoll et al. / Clinical Microbiology and Infection 28 (2022) 1288.e1e1288.e7
of Medical Mycology, Gilead Sciences, GPR Academy Ruesselsheim,
medupdate GmbH, MedMedia, MSD Sharp & Dohme GmbH, Pfizer
Pharma GmbH, Scilink Comunicacio  n Científica SC, and University
Hospital and LMU Munich; participation on an Advisory Board from
Ambu GmbH, Gilead Sciences, Mundipharma Research Limited, and
Pfizer Pharma; A pending patent currently reviewed at the German
Patent and Trade Mark Office; other non-financial interests from
Elsevier, Wiley, and Taylor & Francis online outside the submitted
work. A.A. reports payment or honoraria for lectures, presentations,
speaker's bureaus, manuscript writing, educational events, and
support for attending meetings and/or travel from Gilead and
Pfizer, as well as patents planned, issued, or pending for Pneumo-
cystis PCR and Histoplasma PCR. C.L.F. reports grants from Gilead
Sciences and Astellas Pharma, consulting fees from Merck Sharp
and Dohme and Gilead Sciences, support for attending meetings
and/or travel from Gilead Sciences, and honoraria for lectures/
presentations from Gilead Sciences, Merck Sharp and Dohme,
Pfizer, BioMerieux, F2G and Immy outside the submitted work.
M.K., N.L., H.U., E.S., H.L.V., S.H. and F.F. have nothing to disclose.
This research was funded by W1253-B24 doctoral program
HOROS (C.L.F).
Author contributions
Conceptualization and methodology: C.L.-F. and M.A.K.. Acqui-
sition, analysis and interpretation of data: E.S., N.L., M.A.K., H.L.V.,
P.-M.R., J.S., F.F., P.K., A.A., and R.K.. Statistical analysis: H.U. and
M.A.K.. Supervision: C.L.-F.. Writingeoriginal draft preparation:
M.A.K., C.L.-F.. Writingereview and editing: all authors.
Acknowledgements
The authors would like to thank Andrea Hain, Mona Schrepffer
(Essen), Annika Dietz, and Manja Schmidt (Nuernberg) for their
excellent technical assistance.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.cmi.2022.04.014.
References
[1] Pappas PG, Lionakis MS, Arendrup MC, Ostrosky-Zeichner L, Kullberg BJ.
Invasive candidiasis. Nat Rev Dis Primers 2018;4:18026. https://doi.org/
10.1038/nrdp.2018.26.
[2] Pfaller MA, Diekema DJ, Turnidge JD, Castanheira M, Jones RN. Twenty years of
the SENTRY antifungal surveillance program: results for Candida species from
1997-2016. Open Forum Infect Dis 2019;6:S79e94. https://doi.org/10.1093/
ofid/ofy358.
[3] Ullmann AJ, Aguado JM, Arikan-Akdagli S, Denning DW, Groll AH, Lagrou K,
et al. Diagnosis and management of Aspergillus diseases: executive summary
of the 2017 ESCMID-ECMM-ERS guideline. Clin Microbiol Infect 2018;24
Suppl 1:e1e38. https://doi.org/10.1016/j.cmi.2018.01.002.
[4] Ahmad S, Joseph L, Hagen F, Meis JF, Khan Z. Concomitant occurrence of
itraconazole-resistant and -susceptible strains of Aspergillus fumigatus in
routine cultures. J Antimicrob Chemother 2015;70:412e5. https://doi.org/
10.1093/jac/dku410.
[5] Cornely OA, Hoenigl M, Lass-Flo € rl C, Chen SC-A, Kontoyiannis DP,
Morrissey CO, et al. Defining breakthrough invasive fungal infection-position
paper of the mycoses study group education and research consortium and the
European Confederation of Medical Mycology. Mycoses 2019;62:716e29.
https://doi.org/10.1111/myc.12960.
[6] Jenks JD, Gangneux JP, Schwartz IS, Alastruey-Izquierdo A, Lagrou K,
Thompson Iii GR, et al. Diagnosis of breakthrough fungal infections in the
clinical mycology laboratory: an ECMM consensus statement. J Fungi (Basel)
2020;6:216. https://doi.org/10.3390/jof6040216.
 E, Maufrais C, Sertour N, Palayret M. d'Enfert C, Bougnoux ME. Within-
[7] Sitterle
host genomic diversity of candida albicans in healthy carriers. Sci Rep 2019;9:
2563. https://doi.org/10.1038/s41598-019-38768-4.
1288.e6
[8] Zhang N, Wheeler D, Truglio M, Lazzarini C, Upritchard J, McKinney W, et al.
Multi-locus next-generation sequence typing of DNA extracted from pooled
colonies detects multiple unrelated Candida albicans strains in a significant
proportion of patient samples. Front Microbiol 2018;9:1179. https://doi.org/
10.3389/fmicb.2018.01179.
[9] Ford CB, Funt JM, Abbey D, Issi L, Guiducci C, Martinez DA, et al. The evolution
of drug resistance in clinical isolates of Candida albicans. Elife 2015;4. https://
doi.org/10.7554/eLife.00662. e00662.
[10] The European Committee on Antimicrobial Susceptibility Testing. Method for
the determination of broth dilution minimum inhibitory concentrations of
antifungal agents for yeasts: EUCAST definitive document E.DEF 7.3.2. EUCAST
Antifungal MIC Method for Yeast. Available at:: https://www.eucast.org/
fileadmin/src/media/PDFs/EUCAST_files/AFST/Files/EUCAST_E_Def_7.3.2_
Yeast_testing_definitive_revised_2020.pdf. Accessed January 24, 2022.
[11] bioMe rieux SA. Package Insert - 9305056: D - en - Etest - AFST WW. Available
at: https://techlib.biomerieux.com/wcm/techlib/techlib/documents/docLink/
Package_Insert/35905001-35906000/Package_Insert_-_9305056_-_D_-_en_-_
Etest_-_AFST_WW366969.pdf. Accessed July 18, 2021.
[12] The European Committee on Antimicrobial Susceptibility Testing. Breakpoint
tables for interpretation of MICs for antifungal agents, version 10.0. Available
from:: http://www.eucast.org/astoffungi/clinicalbreakpointsforantifungals/.
Accessed July 18, 2021.
[13] Arendrup MC, Boekhout T, Akova M, Meis JF, Cornely OA, Lortholary O. ESC-
MID and ECMM joint clinical guidelines for the diagnosis and management of
rare invasive yeast infections. Clin Microbiol Infect 2014;20 Suppl 3:76e98.
https://doi.org/10.1111/1469-0691.12360.
[14] Lockhart SR, Iqbal N, Cleveland AA, Farley MM, Harrison LH, Bolden CB, et al.
Species identification and antifungal susceptibility testing of Candida blood-
stream isolates from population-based surveillance studies in two U.S. cities
from 2008 to 2011. J Clin Microbiol 2012;50:3435e42. https://doi.org/
10.1128/JCM.01283-12.
[15] Toth Z, Forg
acs L, Locke JB, Kardos G, Nagy F, Kovacs R, et al. In vitro activity of
rezafungin against common and rare Candida species and Saccharomyces
cerevisiae. J Antimicrob Chemother 2019;74:3505e10. https://doi.org/
10.1093/jac/dkz390.
[16] Espinel-Ingroff A, Canto n E, Pema n J. Antifungal resistance among less prev-
alent Candida non-albicans and other yeasts versus established and under
development agents: a literature review. J Fungi (Basel) 2021;7:24. https://
doi.org/10.3390/jof7010024.
[17] New England BioLabs. Instruction manual for monarch genomic DNA purifi-
cation kit, NEB #T3010S/L: Version 3.0_2/20. Available at:: https://www.neb.
com/-/media/nebus/files/manuals/manualt3010.pdf. Accessed July 18, 2021
[18] Jolley KA, Bray JE, Maiden MCJ. Open-access bacterial population genomics:
BIGSdb software, the PubMLST.org website and their applications. Wellcome
Open Res 2018;3:124. https://doi.org/10.12688/wellcomeopenres.14826.1.
[19] Bougnoux ME, Morand S. d'Enfert C. Usefulness of multilocus sequence typing
for characterization of clinical isolates of Candida albicans. J Clin Microbiol
2002;40:1290e7. https://doi.org/10.1128/JCM.40.4.1290-1297.2002.
[20] Dodgson AR, Pujol C, Denning DW, Soll DR, Fox AJ. Multilocus sequence typing
of Candida glabrata reveals geographically enriched clades. J Clin Microbiol
2003;41:5709e17. https://doi.org/10.1128/JCM.41.12.5709e5717.2003.
[21] Tavanti A, Davidson AD, Johnson EM, Maiden MCJ, Shaw DJ, Gow NAR, et al.
Multilocus sequence typing for differentiation of strains of Candida tropicalis.
J Clin Microbiol 2005;43:5593e600. https://doi.org/10.1128/JCM.43.11.55
93e5600.2005.
[22] Valerio HM, RdCB Weikert-Oliveira. Resende MA de. Differentiation of
Candida species obtained from nosocomial candidemia using RAPD-PCR
technique. Rev Soc Bras Med Trop 2006;39:174e8. https://doi.org/10.1590/
S0037-86822006000200008.
[23] Attia E, Elazzaz W. Genotyping of different clinical Candida albicans isolates
using RAPD analysis. Egypt J Exp Biol (Bot) 2020;16:59e69. https://doi.org/
10.5455/egyjebb.20200225042038.
[24] Arendrup MC, Kahlmeter G, Rodriguez-Tudela JL, Donnelly JP. Breakpoints for
susceptibility testing should not divide wild-type distributions of important
target species. Antimicrob Agents Chemother 2009;53:1628e9. https://
doi.org/10.1128/AAC.01624-08.
[25] Pfaller MA, Castanheira M, Diekema DJ, Messer SA, Moet GJ, Jones RN. Com-
parison of European Committee on Antimicrobial Susceptibility Testing
(EUCAST) and Etest methods with the CLSI broth microdilution method for
echinocandin susceptibility testing of Candida species. J Clin Microbiol
2010;48:1592e9. https://doi.org/10.1128/JCM.02445-09.
[26] Dannaoui E, Paugam A, Develoux M, Chochillon C, Matheron J, Datry A, et al.
Comparison of antifungal MICs for yeasts obtained using the EUCAST method
in a reference laboratory and the Etest in nine different hospital laboratories.
Clin Microbiol Infect 2010;16:863e9. https://doi.org/10.1111/j.1469-
0691.2009.02997.x.
[27] Espinel-Ingroff A, Sasso M, Turnidge J, Arendrup M, Botterel F, Bourgeois N,
et al. Etest ECVs/ECOFFs for detection of resistance in prevalent and three
nonprevalent Candida spp. to triazoles and amphotericin B and Aspergillus
spp. to Caspofungin: further assessment of modal variability. Antimicrob
Agents Chemother 2021;65. https://doi.org/10.1128/AAC.01093-21.
e0109321.
[28] Espinel-Ingroff A, Arendrup M, Canto n E, Cordoba S, Dannaoui E, García-
Rodríguez J, et al. Multicenter study of method-dependent epidemiological
cutoff values for detection of resistance in Candida spp. and Aspergillus spp. to
1288.e7 M.A. Knoll et al. / Clinical Microbiology and Infection 28 (2022) 1288.e1e1288.e7

amphotericin b and echinocandins for the Etest agar diffusion method.
Antimicrob Agents Chemother 2016;61. https://doi.org/10.1128/AAC.01792-
16. e01792e16.
 M, Gangneux JP, Cassaing S, Delhaes L, Fekkar A, Dupont D, et al.
[29] Salse
Multicentre study to determine the Etest epidemiological cut-off values of
antifungal drugs in Candida spp. and Aspergillus fumigatus species complex.
Clin Microbiol Infect 2019;25:1546e52. https://doi.org/10.1016/j.cmi.20
19.04.027.
[30] Ben-Ami R, Zimmerman O, Finn T, Amit S, Novikov A, Wertheimer N, et al.
Heteroresistance to fluconazole is a continuously distributed phenotype
among Candida glabrata clinical strains associated with in vivo persistence.
mBio 2016;7. https://doi.org/10.1128/mBio.00655-16. e00655e16.